code-change
Use for a change to this repository's own code that no domain skill covers: fix a bug, patch a loader, add a flag, refactor a function, make it do X.
Skill by agentchieflou
Chalk Labs
Need help deploying code-change?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/agentchieflou-code-change
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Nov 1, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/code-change
What's inside
- Skill name
code-change- Description
- Use for a change to this repository's own code that no domain skill covers: fix a bug, patch a loader, add a flag, refactor a function, make it do X. Guard first, the smallest diff, tests before and after, commit on the ticket's branch.
- License
- MIT
- Instructions
- 28 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Code change The fallback for ordinary code work. Domain skills come first — a measure is `tmdl-edit`, a slow path is `perf-optimize` — and this skill is what is left: a named change to named code. 1. **Scope.** Name the file(s) and symbol(s) the change touches, and the behaviour that proves it is done, in one line each. Cannot name them from the request → `ad-state ask "<which file / what behaviour>" --want value`, `friction-log` type `missing-info`. STOP. A safe, reversible default settles it (a test file beside the module, an obvious name) → `ad-state ask "<assumption>" --assume "<default>"`, one line, CONTINUE. 2. **Branch** (AGENTS.md rule 16): `git for-each-ref refs/heads --format=%(refname:short)`. A branch carrying the ticket key exists → continue on it if it is checked out, else `ad-state ask` the operator to check it out. STOP. None → `git checkout -b feature/<KEY>-<slug≤4 words>` (untracked: `fix/<slug>`), then `ad-state set branch=<branch>`. 3. **Guard** (AGENTS.md rule 14): `ad-graph status`. Not `approved: current` → invoke `codebase-map`; STOP. Then `ad-test run --snapshot before`. A red suite before you touched anything → `friction-log` type `contract`; STOP. Zero tests collected is not green: `friction-log` type `contract` naming `test_cmd`; STOP. 4. **Edit** the named symbols only (`ad-state set phase=editing` first). The smallest change that does what step 1 says. Never change a public signature the request did not name, never touch a second module "while you are in there". 5. `ad-graph guard`. `ok: false` → `git checkout -- <file>`, print the refused rows. An uncovered node → hand off to `test-cover` for it, then come back here. STOP otherwise. 6. **Prove it.** `ad-test run --snapshot after`, then `ad-test run --compare <before.tsv> <after.tsv>` must be `ok: true`. Add or extend one test that fails without the change and passes with it. A regression → revert the edit, quote the row, STOP. Two failed attempts → `friction-log`; STOP. 7. Commit `fix: <KEY> <what>` or `feat: <KEY> <what>` (untracked: no key). `ad-state set phase=validating`. Print the files changed and the test that proves it. Hand off → `bitbucket-pr`.
From agentchieflou/this-next-please, MIT.
More from agentchieflou/this-next-please
- 0.592score
data-adapter
agentchieflou
How to read and produce data in this workspace.
SkillAFreshness: A★ 1Claude Code +9 - 0.592score
jira-changelog
agentchieflou
Use for field history, sprint reports, committed vs completed points, "when did X change", or anything current Jira values cannot answer.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
confluence-publish
agentchieflou
Use to write findings, runbooks, or work documentation to Confluence.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
file-organize
agentchieflou
Use when a folder of files has to be organized — a delivery of documents to sort before a DPM run, a heap with no structure, 'where should these go', 'file…
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
hive-query
agentchieflou
Use for read-only SQL against Hive (HiveServer2), Impala or Spark SQL (the Spark Thrift Server) on Hadoop.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
friction-log
agentchieflou
Use the moment you second-guess yourself, repeat a tool call, hit ambiguous acceptance criteria, or get two failures in a row.
SkillAFreshness: A★ 1Claude Code +9 - 0.582score
dax-studio-export
agentchieflou
Use to evaluate a DAX query against a deployed model or a running Desktop and get results as TOON, or to export a .vpax model-metrics file.
SkillAFreshness: A★ 1Claude Code +9 - 0.582score
bitbucket-pr
agentchieflou
Use when code or model changes are ready for review — to branch, commit, push, and open a Bitbucket pull request with ad-git pr.
SkillAFreshness: A★ 1Claude Code +9 - 0.579score
code-router
agentchieflou
Domain sub-router for work on a repository's own code: map it, cover it, change it, make it faster, prove nothing broke, tidy the worktree.
SkillAFreshness: A★ 1Claude Code +9 - 0.579score
dpm-router
agentchieflou
Domain sub-router for document pipelines: DPM run handoff, field extraction, Content Understanding analyzers, organizing a folder of documents.
SkillAFreshness: A★ 1Claude Code +9 - 0.579score
jira-comment
agentchieflou
Use to post a comment on a Jira issue without moving it — a finding, a question for the reporter, a status note, the path of an artifact.
SkillAFreshness: A★ 1Claude Code +9 - 0.579score
jira-router
agentchieflou
Domain sub-router for Jira work: triage a key, create a ticket, field history and sprint replay, move an issue.
SkillAFreshness: A★ 1Claude Code +9
Signals
- Stars
- 1
- Forks
- 0
- Open issues
- 61
- Last commit (this folder)
- Oct 3, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from agentchieflou
- 0.592score
jira-changelog
agentchieflou
Use for field history, sprint reports, committed vs completed points, "when did X change", or anything current Jira values cannot answer.
SkillAFreshness: A★ 1Claude Code +9 - 0.592score
data-adapter
agentchieflou
How to read and produce data in this workspace.
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
file-organize
agentchieflou
Use when a folder of files has to be organized — a delivery of documents to sort before a DPM run, a heap with no structure, 'where should these go', 'file…
SkillAFreshness: A★ 1Claude Code +9 - 0.587score
confluence-publish
agentchieflou
Use to write findings, runbooks, or work documentation to Confluence.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
friction-log
agentchieflou
Use the moment you second-guess yourself, repeat a tool call, hit ambiguous acceptance criteria, or get two failures in a row.
SkillAFreshness: A★ 1Claude Code +9 - 0.584score
hive-query
agentchieflou
Use for read-only SQL against Hive (HiveServer2), Impala or Spark SQL (the Spark Thrift Server) on Hadoop.
SkillAFreshness: A★ 1Claude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get code-change running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
code-change skill. Agent Mart. https://agentmart.dev/skills/agentchieflou-code-change (last updated 2026-10-24)